Lee Brady's Recording Blog

GCN and Arable - Some Thoughts

 

[this information was originally in the previous post, but I put it here to unclutter the status assessment]

 

Mike Phillips dropped me an email to query the arable. So I thought it worthwhile to post my thoughts. 

A problem we have is that there is little available quantitative data on terrestrial habitat use. Since most gcn records are collected from ponds, the model is therefore scoring terrestrial landscape features based on their suitability as locations for gcn ponds. 

A grassland location scores much better than a woodland location (the latter barely positive). Arable is in the middle. Builtup locations and locations with conifer typically score negatively (ie. newts are infrequently recorded in these areas). 

One possible modification to the status assessment would be to provide a range of values. For terrestrial habitat, the lower range would exclude arable, the maximum would include it. However, I am reminded of mitigation projects that capture many gcn in arable fields.

Since the model used for the Kent Status Assessment is summing the % habitat cover in only those squares where gcn presence is predicted, it already controls for arable areas with high % cover where gcn are likely to be absent.

Looking at the model outputs, 1 km squares with very extensive % cover of arable cannot overcome the low score generated if the square has few ponds. So 1 km squares with very high (>90%) arable cover and 3 or less ponds typically generate a Low score in terms of their suitability for gcn.

In comparison, 1 km squares with few ponds and high % cover of improved grassland will usually generate a Medium score in terms of their suitability for gcn. 

Of course the overall score depends on the mix of landscape features that are present, with pond density being the main positive influence (by a long way).

Expressed another way. A 1 km square located in the range of gcn with very high arable % cover will need more ponds created than a square with a lower % cover of grassland to achieve the same suitability score. 

For interest, the highest scoring 1 km square with low pond density (3 ponds in this case) has the following mix of terrestrial habitat classifications:

Improved Grassland = 54.7%
Arable = 39%
Broadleaf = 0
Conifer = 0
Semi-Natural = 2.6%
Upland = 3.7%
Builtup = 0

For low pond density squares with an overall score of High, the square with the highest arable % cover has the following composition:

Improved Grassland = 22%
Arable = 63.4%
Broadleaf = 0
Conifer = 0
Semi-Natural = 6.6%
Upland = 7.8%
Builtup = 0

(there are 3 ponds in this square and gcn are confirmed present)

 

Last comment. Providing there are a reasonable number of ponds present, the presence of arable will not prevent pond occupancy by gcn - providing the square is actually within their range.   

 

Great Crested Newt Kent Status Assessment 2017

In advance of Natural England publishing a conservation strategy for great crested newts in Kent I thought it would be timely to publish my own status assessment using records that are available up to March 2017.

The term 'Kent' describes an area that is something of a moving target - due to changes in political boundaries. For the purposes of this analysis I will be including all areas that are defined within the boundary of the two Vice Counties: East Kent and West Kent. This should be taken into consideration when making comparisons to any other data that may be published by others in the future. 

 

 

The status assessment will use the broad categories identified by JNCC in their description of Favourable Conservation Status. These include:

Range

Population

Habitat &

Future Prospects

 

Future Prospects is likely to include factors that may positively or negatively influence the future conservation status of gcn. One such factor is climate change. I will leave Future Prospects for a future discussion. 

 

Range

Range simply describes the maximum expected occurrence of a species within the area of interest. It can be defined in several different ways and many assessments simply use basic Distribution maps. Here is a distribution map for gcn in Kent.

 

It is worth noting that only a very small number of gcn populations occur in that area of West Kent that is now considered to form part of Greater London.  

 

For the current assessment of range I have chosen to use a model that combines recorded distribution with Landscape Habitat Suitability. 

 

No. of 1 km squares in Kent (i.e. total area) = 4456

No. of 1 km squares predicted to be occupied by gcn = 2613 km

 

Of course, not all areas offer the same suitability and the prediction therefore determines range based on three levels of predicted occupancy (low, medium & high). The figures quoted above include only 1 km squares with scores of medium or high.

The predicted range of gcn can be visualised as follows.

 

 

Population

It is unrealistic to expect surveyors to count the actual number of gcn in Kent. At individual ponds, populations can be assessed by using simplistic methods that provide relative counts (eg. the methods currently required by NE for gcn licence applications), or more complicated statistical methods that control for the detectability of newts in a pond (eg. capture - recapture).

But even simple relative counts require significant effort and are impractical for estimating the overall Kent population. I have therefore chosen to represent population as the number of occupied ponds. GCN can occupy ditches and such waterbodies are not included within the current assessment.  

Even so, there are a lot of ponds in Kent...

 

Ponds stored on the database are subject to constant review. New ponds are discovered and added, old ponds may be found to no longer exist. As of March 2017, there are 17637 ponds included within the Kent database...

Of these ponds, 17206 are found in 1 km squares where we predict gcn are likely to occur. 

This is very clearly most of the ponds! But remember that the habitat that most influences gcn presence in Kent is pond density, so this should not really be a surprise.  

Previous work has investigated the relationship between individual pond habitat suitability and rate of occupancy of those ponds by gcn. Using these approaches, we have predicted that gcn will be found in 47.1% of available ponds. We can therefore estimate the total number of ponds in Kent that are occupied by gcn:

 

0.471 x 17206 = 8104 occupied ponds (or populations) 

 

Of course, not all of these ponds will be high quality and considered suitable for breeding. By restricting our assessment to only those ponds that are likely to score as good or excellent (26.8%), we can estimate the number of ponds in Kent that are considered suitable for breeding by gcn:

 

0.268 x 17206 = 4611 breeding ponds (or breeding populations)

 

Habitat

GCN Habitat can be considered as either terrestrial or aquatic. For aquatic habitat, we may wish to simply consider the number of ponds (17206) or the area of those ponds. The relationship between gcn occupancy and pond area is unclear. GCN appear to favour medium sized ponds and tend to avoid very small and very large ponds (possibly because of pond permanence and likely presence of fish respectively). But there are many exceptions. 

Since pond area is a value recorded on the database, we can sum the total area of ponds known to be present in the predicted range of gcn:

 

Aquatic Habitat = 7311780 square meters = 731.18 Ha = 7.31 square km

 

For terrestrial habitat, we can examine our range model and identify non-aquatic landscape features that are considered to positively influence the likely presence of gcn. These include:

 

improved grassland

arable (yes, that's correct - arable!) - see next post for some discussion

broadleaf woodland

upland

 

By calculating the total area of these landscape features, we can estimate the extent of suitable terrestrial habitat within the range of gcn:

 

Terrestrial Habitat = 1844  square km

 

We may consider the total area of suitable habitat available to gcn as the sum of aquatic and terrestrial habitat values:

 

Total Habitat = 1851 square km